In a prospective longitudinal study, it is hypothesized that the built environment's walkability will impact the social environment's collective efficacy, and that these environmental variables together will influence physical activity. Controlling for diet, physical activity is hypothesized to predict adiposity, inflammation, and insulin resistance;and these in turn will predict progression in metabolic syndrome indicators. While these relationships have been tested separately, this application proposes an integrated and interdisciplinary test of the relationship among these variables. Hispanic immigrants are selected for this study because Hispanics have disproportionately high rates of diabetes. Moreover, for very recent immigrants, weight increases rapidly as a function of time in the US. The study proposes to explore trajectories of cultural change - and their impact on diet and physical activity - as a mechanism by which 'time in the US'may bring about its deleterious health outcome for this population. A sample is proposed that is uniquely suited to investigate this phenomenon in Hispanics, as well as the impact of the built environment on physical activity and health outcomes: Cuban immigrants within 90 days of arrival to the US. This sample addresses in part the self- selection bias that characterizes most built environment studies, because, when these individuals arrive in Miami-Dade County, they typically move to a neighborhood not of their choosing. Therefore, an ideal natural experiment occurs in which a naive sample in relation to non-walkable environments is distributed across a variety of walkability conditions. Participants will be 390, 30-45 year-old recent immigrants who do not meet criteria for metabolic syndrome. Built environment's walkability will be assessed through self-reports, and objectively using the .25 mile radius surrounding the participant's residence. Participants will be assessed on self-reported measures of collective efficacy, physical activity, diet, and biculturation, and objective measures of physical activity, adiposity, inflammation, insulin resistance, and metabolic syndrome indicators. Variables in the primary analyses will be continuously distributed. Growth curves will be estimated using growth curve methodology in a structural equations framework using Mplus 3.